Image sensors are widely used in a variety of consumer electronic devices, including digital cameras and cellular telephones with built-in digital cameras. An image sensor may comprise a matrix of sensor elements. If one or more sensor elements are defective or occluded, this condition may degrade image quality.
Sensor elements may include “bright defects” and “dark defects,” perhaps causing them to be unusable. Some bright and dark defects may be of a “stuck pixel” variety. That is, an output signal from a stuck sensor element may register a fixed signal level regardless of the brightness of light incident to the element. The terms “defective element” and “unusable element” are used interchangeably herein. “Brightness of light” may hereinafter be referred to using the more precise term “illuminance.” Units of lux, or lumens per square meter, may be used as a measure of illuminance.
An image sensor may convert light to analog sensor element output signals. An analog to digital converter (ADC) may convert the sensor element output signals to a digital format. A least significant bit (LSB) associated with the ADC may establish a granularity with which illuminance sensed by the sensor element may be measured. Thus, in the field of digital imaging, illuminance may also be expressed in LSB units.
A dark defect may also result from a dust particle lodged against the sensor element, blocking some or all of the light that might otherwise impinge on the element. If all light is blocked, the sensor element output signal may remain fixed as the illuminance incident to the sensor element changes. If the incident light is only partially blocked, and the partially-blocked sensor element is not stuck, the element output signal may vary as the incident illuminance varies. In the latter case, however, the output signal may not be as large as if the sensor element were not partially blocked.
Some methods are available to identify unusable sensor elements at a production facility following fabrication. For example, each production sensor may be tested and calibrated under controlled conditions. Such methods may incur substantial cost, and may fail to identify dark defects resulting from dust particles.
Consider, for example, a cellular telephone with a built-in camera. The camera sensor may have been tested and calibrated at the semiconductor manufacturing facility, following fabrication and before shipment to a cellular telephone manufacturer. During camera module assembly, a dust particle inside the camera module may break free and land on the surface of one or more sensor elements. The resulting dark defect may subsequently manifest itself as a cluster of dark pixels on captured images. Thus, there is a need to identify and rectify such defects.